Tip jar

If you like CaB and wish to support it, you can use PayPal or KoFi. Thank you, and I hope you continue to enjoy the site - Neil.

Buy Me a Coffee at ko-fi.com

Support CaB

Recent

Welcome to Cook'd and Bomb'd. Please login or sign up.

March 29, 2024, 08:14:15 AM

Login with username, password and session length

[Muso] Music production.

Started by small_world, September 26, 2011, 06:38:35 AM

Previous topic - Next topic

small_world

That 'polished sound'.
How the hell do you get it? And I know some of you do, I've heard it on things posted up on here.
So, what do you do to get it sounding all tight and closed, all nice and airey, all real and beautiful?

This is something I've been working on for a little while.
http://soundcloud.com/torslanda/volcano-day-fin

I'ts not the kind of thing I'd imagine most of you would like. But, content aside, what do I need to do to give it 'that sound'?

NoSleep

Bloody fuck; 328mb for one 3m44s track? A whole CD album is only 650mb! That's got to be a 64-bit file. You need to render it at 24-bit or 16-bit. Not heard it yet, as a result; but as someone who works as a professional recording engineer, I'd say "slick" is overrated. Everything should be wrought for maximum emotion, not to some lofty standard of "slick". You know you're getting there when the head starts to nod automatically.

Dusty Gozongas

Ah. Mastering. It's a bit of a dark art is that. Personally I rarely bother as my own stuff is generally for personal consumption although it's quite pleasing to the (my) ear when I do.

Depending on your software/hardware, you might find a Google or YouTube search helps in finding hints and tips specific to your set-up.

Wikipedia gives some general background info. NoSleep is totally correct in every respect though :-)

Puffin Chunks

EQ.

That's your short answer. And there are no real shortcuts to getting there. You can use spectrum analysisers and the like to visualise where your sounds are sitting in the mix, but ultimately it comes down your ears and experience. Try and get an idea in your head about where different sounds should fit in the audio spectrum and fit them there. There will always be overlap, but in short - lots of sounds competing for the same frequency space = muddy mix.

A lot of 'pro' music is compressed to hell. This is a bad thing, so avoid it! Dynamic range is important, but apparently not in modern music. However compression without EQ will still sound bad and probably a lot worse.

Oh, and in addition to this, boosting high frequencies (using an exciter can help to get that 'crisp, shiny' sound. But again, it is an after effect that should be used in addition to EQ not instead of. Also, too much highs sounds tinny and painful to the listener, and boosting the highs on all sounds just sounds terrible, so use sparingly.

NoSleep

I've sorted an mp3 of your track (the file on Soundcloud didn't even start for me[nb]Running on sail here; an old G4; until my "new" G5 arrives.[/nb], so I downloaded it).

http://dl.dropbox.com/u/318457/Volcano%20Day.mp3

Synth bass line and drum machine with guitar performed (with overdubs/drop-ins) over them? The drums are the only real problem for me. Better sounds could be found to use, and they could do with being tweaked up in volume a bit. The little hihat pattern that sticks out in the chorus could be made to fit in better, too - some of its notes should be accented and others set back; just to make it sound less machine-like. I'll have another listen later, but first impressions are always the best (or leave a while to relisten).

The Masked Unit

I've been meaning to ask some mixing questions of our experts here, so this is probably as good a place as any.

How do people tend to approach a final mix? Bearing in mind that everything I do is synth and sample based with no proper instrumentation, and I mix "inside the box" rather than through a desk, what I tend to do once I've got my arrangement sorted is to pull all the faders down and start from scratch, starting by getting the kick drums, bass and low frequency percussion like toms etc working well together. I then do the same with the high frequency parts, starting with the hats, and finally slot everything else in. I find that it's good to do that with the low frequency stuff and also the free up some space for pads and other synth parts that are competing for the same part of the spectrum, but the downside is that it often damages what was good about the track - the way it grooves along, basically. Any tips for a halfway house?


NoSleep

I wouldn't separate things on account of their frequency. Stick all the drums together, as well as the bass, including the hi-hats, shakers and bells, etc and get them all grooving together. Same thing with the rest of the music; stick them all together first of all. Same with vocals. Then that will leave any extras, like samples, scratches and sound fx. So, four basic groups. Maybe send them to four stereo submix groups. By treating them this way you can adjust the overall volume of the final mix (assuming you're working in digital) by tweaking the submixes, rather than having to go back to individual faders, if you're going into the red (ie, over 0db, the absolute ceiling of digital recordings[nb]Although most music programs keep a little bit (6db) of ceiling up their sleeve.[/nb]).

Bear in mind that often the bass line will suit being grouped together with the drums if it's a standalone riff, but sometimes the bass will be more of a "warmer" for the music group and might benefit being grouped with them instead.

Shoulders?-Stomach!

I couldn't tell you anything over NoSleep, but from hearing the track I'd make the following comments:

- The guitar line is quite singular so because it's constantly going on it emphasises the absence of much in the background. It needs something else, especially higher up, to give it a stronger sense of harmony, progression and emotion. If that something had a rhythmical element too, I think that would help.

- Some of the mini-sections do sound fairly polished- the ones with repeated melodies sound fine.

-The actual structure of the song sounds like a block of sound to be used as background fodder rather than a song whose purpose is to establish a feel, develop on it, go into different areas etc.

-The choice of drums and bass is pretty good. Some of the fills and little sections are great. Overall, I don't think the stock drums are strong enough to sustain the whole track, and perhaps at points don't sound like they belong in the same musical space as the guitar element. Some wide, but subtly mixed reverb and more bass in the equaliser might help you.

-The guitar sounds very nice and most of the timings are great and that compliments the sense of professionalism. However, as that's the best sounding element of the piece, I think it's too singular and it goes overboard. I wouldn't say it's that far away from some of the more muzaky slightly-half arsed background rock library music you hear used. I really don't mean that as a bad thing.

Shoulders?-Stomach!

By the way, I hope the criticism is acceptable.

I tend to agree with NoSleep- whatever in a particular piece makes it resonate is more important than achieving an arbitrary goal. Plenty of albums I've heard have lost elements that were there in live performances and demos through being endlessly played with.

Lots of modern pop music has very busy production, but often not to that great effect, because the core things- the singing, the structure and the 'sense'- that amazing ambience of a group of things bouncing off each other organically are poor or missing. Also, the desire for the production to fill every bit of space in a song with something I think makes for attritional listening over the course of a whole album.

This isn't to say to leave things that sound bad. It's more a case of identifying when you have added something positive, and when things need changing for the good of the track.

small_world

Quote from: Shoulders?-Stomach! on September 26, 2011, 06:01:34 PM
By the way, I hope the criticism is acceptable.

Yeah, totally.
Really glad people have actually bothered to listen, but the feedback is great.
And, as mentioned I know it's not the kind of thing that would go down great here (well after reading the 'Vai/noodling guitar thread' from ages back) But the criticisms have been constructive without going in to actual 'type' of music it is.

While, obviously I like that kind of rock guitar (yes, almost 'library' ) music, I would like it to be quite interesting. So, I am going to go back and work on a few of the other instruments. It was something I had meant to work on, particularly adding something to the drums, to help them stand out.

The 'drum machine sound' is again, something I was kind of aiming for.
Like this. http://www.youtube.com/watch?v=jLJtO_iRbA8
But I take the point that it should have SOME interesting stuff to it.

Sorry about the size. This is the first time I've put something online, so I'm going to get this sorted tonight, re-uploaded with a smaller file size.

By the way, if you're reading this NoSleep should my CaB album mix be the size this thing was?



On pop music. Jeez, some of that shit is so over the top, way too many tracks, and a lot of it sounds just way too busy.
I can't actually think of anything where it works. It's just so un-dynamic. Just a thrust into the song, and a load of noise until it ends. So many things going on at the same time, and no room to breathe.

But yeah. I'll work a bit more on this, get a few others up too. And thanks.

NoSleep

Quote from: small_world on September 26, 2011, 06:59:27 PM
Sorry about the size. This is the first time I've put something online, so I'm going to get this sorted tonight, re-uploaded with a smaller file size.

By the way, if you're reading this NoSleep should my CaB album mix be the size this thing was?


I'm still not sure how you managed to make it that big, although REAPER can chuck out a 64-bit version of a mix as one of the options. You're working in Ableton, aren't you? If you chuck out a 24-bit mix, that should fine (must be an option) and adding a bit of dithering and noise-shaping would be a bonus.

alan nagsworth

Quote from: Puffin Chunks on September 26, 2011, 07:45:13 AM
EQ.

Innit I agree. Past that, I just apply effects and filters and use EQs to achieve the desired effect. If the vocals are too muffled just put a bit of light distortion on to bring up the hiss of the accentuation more. If the cymbals are too tinny, play about with the high end on the EQs.

If making music has taught me anything it's that you shouldn't work to any desired standard past your own. Enjoy yourself and even if it means adding a million layers in an unprofesional medium, if it brings out your desired sound then roll with it. Don't get too caught up in it initially because you will overwork yourself and that's the kind of pressure that an artist having fun simply does not need.

I lstened to your song and I really liked it. Get some KICK DRUMZ in there though, man! And the other percussion I think needs to be higher/louder in the mix but with less high-end on the EQ maybe? Other than that, I thought it was cool as fuck.

I might not be the best person to ask though as I'm a big Ween fan and your song reminded me of Ween's brown sound aesthetic merged with a avant-virtuoso Mattias Eklundh style, and I just basically thought it was cool as fuck. It also reminded me of old video games. Right on!

The Masked Unit

Quote from: NoSleep on September 26, 2011, 05:21:40 PM
I wouldn't separate things on account of their frequency. Stick all the drums together, as well as the bass, including the hi-hats, shakers and bells, etc and get them all grooving together. Same thing with the rest of the music; stick them all together first of all. Same with vocals. Then that will leave any extras, like samples, scratches and sound fx. So, four basic groups. Maybe send them to four stereo submix groups. By treating them this way you can adjust the overall volume of the final mix (assuming you're working in digital) by tweaking the submixes, rather than having to go back to individual faders, if you're going into the red (ie, over 0db, the absolute ceiling of digital recordings[nb]Although most music programs keep a little bit (6db) of ceiling up their sleeve.[/nb]).

Bear in mind that often the bass line will suit being grouped together with the drums if it's a standalone riff, but sometimes the bass will be more of a "warmer" for the music group and might benefit being grouped with them instead.

Interesting stuff, I'll give that a go.

glitch

Quote from: Puffin Chunks on September 26, 2011, 07:45:13 AM
Try and get an idea in your head about where different sounds should fit in the audio spectrum and fit them there. There will always be overlap, but in short - lots of sounds competing for the same frequency space = muddy mix.

This might help for that.

I started a Wiki ages ago where I'd copy/paste snippets/tips I found in threads like this and books etc. It's still bare and I've not used it in a while but would people be interested in access to it and try to collaborate on putting together some decent production help?

Shoulders?-Stomach!

A pity so much stuff takes that frequency map as an invitation to fill the gaps, rather than producing sensibly.

jake thunder

Quickly:

1) Send groups (drums, keys etc...) of instruments to busses. Compress / EQ / Whatever those busses if you want
2) Echo / Delay / Reverb - Don't use more than one type of each. Use FX sends to apply echo to more than one track. Use convolution plugins.
3) EQ - Live instruments don't really need EQing unless you fucked up whilst recording them or want a specific EQ sound
4) Bass - Don't increase the low frequencies on the EQ too much. Just turn up your bass instrument in volume
5) Panning - If your instrument is panned, send the echo to other channel
6) Use "fake stereo" FX to make stuff sound good - PSP Pseudo Stereo sounds good. Looks shit. Sounds good.
7) Don't use distortion on vocals unless you want to be a cunt
8) Mastering compression isn't a bad thing. It suits electronic music.

jake thunder

Oh yeah:

9) Balls to fitting sounds into the audio spectrum. That has nothing to do with music. Most real instruments have a large frequency range. Humans are quite capable of listening to both a guitar and a piano at the same time.

glitch

Quote from: jake thunder on September 29, 2011, 02:07:41 PM
Oh yeah:

9) Balls to fitting sounds into the audio spectrum. That has nothing to do with music. Most real instruments have a large frequency range. Humans are quite capable of listening to both a guitar and a piano at the same time.

Eh, I find that drums sound better if you carve them out, e.g. LPF on the kick, then HPF the snares at the same cut-off. Helps separate them a lot.

Also subtractive EQ is better than additive.

Shoulders?-Stomach!

Surely finding the right drum sound, particularly your basic kick or bass drum is one of the easier tasks in music. That would reduce the time spent arsing around with EQ (or 'equaliser' as I will keep calling it).

Most of the time I find myself using that is when I import samples from elsewhere and you get extra/unwanted noise that needs removing, or the sound you were after in the pieces is too faint or muddy, and with microphone recordings.


NoSleep

#19
Quote from: jake thunder on September 29, 2011, 02:07:41 PM
Oh yeah:

9) Balls to fitting sounds into the audio spectrum. That has nothing to do with music. Most real instruments have a large frequency range. Humans are quite capable of listening to both a guitar and a piano at the same time.

All you need to do is pan them, really, and then they can co-exist nicely.[nb]Assuming here that they are sharing the same frequency range.[/nb]

jake thunder

Quote from: NoSleep on September 29, 2011, 06:12:00 PM
All you need to do is pan them, really, and then they can co-exist nicely.[nb]Assuming here that they are sharing the same frequency range.[/nb]

You don't need to pan shit.

NoSleep

Quote from: jake thunder on September 29, 2011, 07:04:17 PM
You don't need to pan shit.

And yet you need to use "fake stereo"? Strange philosophy.

glitch

Quote from: NoSleep on September 29, 2011, 07:08:49 PM
And yet you need to use "fake stereo"? Strange philosophy.

Got to agree with this. Not saying I'm a production expert but panning can really help separate sounds and provide "depth" to the track.

Aside from on the drums and bass, which should be dead centre and is why you need to use filters/EQ to separate them.

NoSleep

#23
I wouldn't exclude the possibility of any number of approaches. It's quite possible for a piano and guitar to sit happily together, but if there is a problem discerning what each is doing (and you desire that difference), then panning will certainly help, even just a little.

I would say that the use of eq on the drums and bass is not because they're together in the centre of the mix as, usually it's only the kick snare and bass sharing the same position. The kick & snare (in most styles) rarely overlap & Toms, cymbals, hi-hats, shakers etc can be panned about. You're not even obliged to keep the kick, snare & bass to the exact centre. The traditional two reasons for sticking them there were that, (1) in the days of vinyl, panning the bass or kick too far to the left or right would make the stylus jump out the groove too easily & (2) Before the days of custom sound systems in dance halls it was important to have the the beat, at least, centred so that, if the left and right speaker stacks were chucked to either end of a hall, the beat of the song was uniform throughout the dance floor. Many dance music mixes are a kind of "virtual mono", where there may only be peripheral sounds coming from extreme left & right positions in the mix; usually stuff like echoes, reverbs and chorus effects (to give it a "surround" effect) as well as less important instruments.

That only leaves the relationship between the bassline and the kick drum to sort out. Best way I've found to deal with that is to boost a narrow band on the kick (somewhere between 55-100Hz) so that it's not taking up a wide bandwidth, then use it to "duck" the bassline a little, using either a noise gate (my preference if you have a noise gate[nb]Working in reverse; turning down the bass when the sidechained kick hits.[/nb] that will allow this) or a compressor. This way you don't get the sludginess when the kick and the bass play together and you can turn them both up higher in the mix (if you want). Done properly you can't notice the volume decrease when the kick hits on the bass; it just makes them both sound punchier.

If you're not producing dance music, nor expecting to have a release on vinyl you have more freedom to pan a bit more.

The Masked Unit

This stuff is gold, NoSleep. Thanks.